List your company

Factuality Detection on HotpotQA (test)

73.3AUROCDynHD
50.9456.74562.5568.355Mar 17, 2026

Evaluation results

MethodLinks
DynHDSource Dataset=TriviaQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0373.3
TSVSource Dataset=TriviaQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0365.5
TraceDetSource Dataset=CSQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0365
DynHDSource Dataset=CSQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0364.9
TSVSource Dataset=CSQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0359.1
TraceDetSource Dataset=TriviaQA, Method Category=Trajectory-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0357.4
CCSSource Dataset=CSQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0354
CCSSource Dataset=TriviaQA, Method Category=Latent-based, Evaluation Protocol=Zero-shot cross-task generalization2026.0351.8